Cascading Modes And Protect Strategies Of Interdependent Networks Base On Edges

A number of critical infrastructure networks are becoming increasingly dependent on each other in real life, such as water grid, transportation network, power grid, Internet and so on. In that case, one point fails in a network will lead to other nodes disabled that are coupled in other networks, and then trigger the spread of failures in interdependent networks, ultimatiely result in a substaintial part of nodes’breakdown, and even the collapse of the whole interdepentdent networks. In response to the catastrophic cascade of failures in critical infrastructures that are coupled together, timely, accurate and targeted rapidly response is the crux of controlling catastrophic spread and reducing socie-economic loss, and also becomes a great challenge in management practice of infrastructure networks. For this purpose, the article mainly pays attention to the cascading failures in interdependent networks, and aims at controlling the spread of cascading failures in interdependent networks effectively, besides, we proposes two cascading modes based on interdependent networks from two perspectives-static and dynamic, explores the cascading dynamic process in interdependent networks base on the effect of edges, additionally, puts forward global and local protection strategies about preventing catastrophic cascade of failures, which could be applied in the field of management. The main contents of this article are as follows:(1) Cascade models of edges in interdependent networks without load are studied. Based on analyzing the dynamics mechanism of cascading failures in coupled networks, the cascading model on the edge is proposed. To better compare the effect of link patterns between coupled networks on the cascading propagation, the correlations between three link patterns, the network topology, and the robustness of coupled networks against cascading failures. By several measures, the optimal structure is obtained. Moreover, we find that, the more similar the coupled networks, the stronger its robustness. In addition, from coupled patterns, network models, and the average degree, the protection strategies in our model are given.(2) Cascade models in interdependent networks with load are studied. Considering the effect of the edge weight and the coupled strength on the load, a new method to define the initial load on an edge is proposed, a cascading load model with tunable parameters is presented. To improve the robustness of interdependent networks against cascading failures, according to some measures, the correlation between the robustness of interdependent networks with the different coupled patterns and some parameters in our model is analyzed from several perspectives. According to several measures, the robustness of coupled networks with different link patterns subject to attacks of the edges with the higher and the lower load is analysised. Finally, by optimizing the link patterns between two networks, the optimized method to enhance the robustness is studied,(3) Case analysis of cascading models. Owing to the fact that power network data in our country is hard to obtain and numerous studies about cascading failures are based on American weastern power grid, therefore, we select the electrical power grid of the western United States as the research subject, coping strategies to prevent cascading failures in real power grid is explored. On the basis of static and dynamic cascading models in interdependent networks, we explore the cascading phenomenon on interdependent networks of power grid from two perspectives, which refer global robustness and the differences between two attacks, and present effective protect strategies. These works will be of both theoretical and practical significance in guiding the development of China power grid, providing reference for improvement of other interdependent networks and so on.The cascade models of edges in interdependent networks we proposed in this article can enrich the analysis of cascading contents. Both the protection strategies and the optimal methods we suggested which can avoid and weaken the disastrous damage induced by cascading failures, could be applied to real-management, and it has vital practical significance in improving the safety and reliability of networks.